In this paper, we propose a novel myopic sensor scheduling methodology for tracking a target moving through a network of energy-constrained acoustic sensors. Specifically, we address the problem of activating the mini...
详细信息
In this paper, we propose a novel myopic sensor scheduling methodology for tracking a target moving through a network of energy-constrained acoustic sensors. Specifically, we address the problem of activating the minimum-energy combination of sensors in a network that maintains a desired squared-error accuracy in the target's position estimate. We first formulate the scheduling problem as a binary (0-1) nonlinear programming (NLP) problem. Using a linearization technique, we then convert the 0-1 NLP problem into a 0-1 mixed integer programming (MIP) problem. We solve the reformulated 0-1 MIP problem using a linear programming relaxation based branch-and-bound technique. We demonstrate through Monte Carlo simulations that our proposed MIP scheduling method is very computational efficient as we can find optimal solutions to scheduling problems involving 50-60 sensors with processing time in the order of seconds
Users and developers of grid applications have access to increasing numbers of resources. While more resources generally mean higher capabilities for an application, they also raise the issue of application scheduling...
详细信息
ISBN:
(纸本)9781424400546
Users and developers of grid applications have access to increasing numbers of resources. While more resources generally mean higher capabilities for an application, they also raise the issue of application scheduling scalability. First, even polynomial time scheduling heuristics may take a prohibitively long time to compute a schedule. Second, and perhaps more critical, it may not be possible to gather all the resource information needed by a scheduling algorithm in a scalable manner. Our application focus is scientific workflows, which can be represented as directed acyclic graphs (DAGs). Our claim is that, in future resource-rich environments, simple scheduling algorithms may be sufficient to achieve good workflow performances. We introduce a scalable scheduling approach that uses a resource abstraction called a virtual grid (VG). Our simulations of a range of typical DAG structures and resources demonstrate that a simple greedy scheduling heuristic combined with the virtual grid abstraction is as effective and more scalable than more complex heuristic DAG scheduling algorithms on large-scale platforms
Input-queued switch architecture has become attractive for implementing high performance routers because the switching fabric and the buffer need not run at a multiple of input link's rate. It is challenging to pr...
详细信息
Input-queued switch architecture has become attractive for implementing high performance routers because the switching fabric and the buffer need not run at a multiple of input link's rate. It is challenging to provide a scheduling technique that is both highly efficient and fair in resource allocation. We propose an iterative cell-based fair scheduling (iCFS) scheme, based on SLIP that support fair bandwidth distribution among flows and achieves asymptotically 100 percent throughput
In order to efficiently utilize available grid resources and promptly complete tasks assigned to the grid, providing a suitable job scheduling strategy for the grid computing is necessary. Lots of grid scheduling algo...
详细信息
In order to efficiently utilize available grid resources and promptly complete tasks assigned to the grid, providing a suitable job scheduling strategy for the grid computing is necessary. Lots of grid scheduling algorithms have already been developed, and some of them are used to schedule independent coarse-grained tasks. Those algorithms don't adapt very well to the grid tasks that are submitted continuously and randomly. Besides, they mostly need a prediction system to provide the prediction information about the processor utilization and the task workloads. This paper proposes an adaptive grid scheduling system for high-throughput applications. Firstly, a grid scheduling model is adopted to represent the performance of processors, the task workloads, and the schedules. Then we develop a scheduling algorithm that doesn't need any prediction information and can adapt to the grid environment. Finally, the scheduling system combines the proposed algorithm with the best of scheduling algorithms that need the prediction information. According to the accuracy of the prediction system in the grid, the system selects the proper strategy to schedule tasks. A prototype of this model is developed and tested with several experiments. The experimental results of the simulation show that the proposed scheduling system is able to perform scheduling well in the grid environment.
To provide performance guarantee for real-time scheduling applications working in open and unpredictable environments, a model predictive control approach is applied to regulate multiple performance metrics of soft re...
详细信息
To provide performance guarantee for real-time scheduling applications working in open and unpredictable environments, a model predictive control approach is applied to regulate multiple performance metrics of soft real-time scheduling systems with uncertain task set characteristic. Utilization and deadline miss control of real-time scheduling is modeled as a multivariable optimization problem. A single input multi-output predictive control framework and algorithm is presented based on a standard quadratic programming solver and a difference equation model that describes soft real-time scheduling systems. The simulation results demonstrate that predictive control scheduling can regulate the utilization and the deadline miss of real-time scheduling systems and show robustness when the actual execution time of tasks deviate from design value at runtime
This paper presents a new operation chaining reconfigurable scheduling algorithm (CRS) based on list scheduling that maximizes instruction level parallelism available in distributed high performance instruction cell b...
详细信息
ISBN:
(纸本)9783981080100
This paper presents a new operation chaining reconfigurable scheduling algorithm (CRS) based on list scheduling that maximizes instruction level parallelism available in distributed high performance instruction cell based reconfigurable systems. Unlike other typical scheduling methods, it considers the placement and routing effect, register assignment and advanced operation chaining compilation technique to generate higher performance scheduled code. The effectiveness of this approach is demonstrated here using a recently developed industrial distributed reconfigurable instruction cell based architecture [Lee,2003]. The results show that schedules using this approach achieve equivalent throughput to VLIW architectures but at much lower power consumption
Sensor networks opened new opportunities to monitor the environment. In order to retrieve the desired data, sensors are usually organized into a hierarchy and synchronize when transmitting the data towards the base st...
详细信息
Sensor networks opened new opportunities to monitor the environment. In order to retrieve the desired data, sensors are usually organized into a hierarchy and synchronize when transmitting the data towards the base station. Many scheduling schemes have been proposed with the goal of allowing sensors to sleep as much as possible and ultimately save energy. In this paper, we propose two new scheduling algorithms that assign predefined slots to each sensor. These algorithms are distributed, need very little global information and do not need knowledge about the location of sensors or the network topology. As others, we also assume that loose clock synchronization is available. The experimental results confirm our expectations. They show a significant reduction in the average time awake per node of at least three times compared to more traditional routing protocols like TAG
The use of meta-schedulers for resource management in large-scale distributed systems often leads to a hierarchy of schedulers. In this paper, we discuss why existing meta-scheduling hierarchies are sometimes not suff...
详细信息
ISBN:
(纸本)9781424430116
The use of meta-schedulers for resource management in large-scale distributed systems often leads to a hierarchy of schedulers. In this paper, we discuss why existing meta-scheduling hierarchies are sometimes not sufficient for Grid systems due to their inability to re-organise jobs already scheduled locally. Such a job re-organisation is required to adapt to evolving loads which are common in heavily used Grid infrastructures. We propose a peer-topeer scheduling model and evaluate it using case studies and mathematical modelling. We detail the DIANA (Data Intensive and Network Aware) scheduling algorithm and its queue management system for coping with the load distribution and for supporting bulk job scheduling. We demonstrate that such a system is beneficial for dynamic, distributed and self-organizing resource management and can assist in optimizing load or job distribution in complex Grid infrastructures.
The relationship between the performance cost and the processor utilization of the real-time control tasks in different control phase was investigated, further more, a feedback scheduling system based on linear assign...
详细信息
The relationship between the performance cost and the processor utilization of the real-time control tasks in different control phase was investigated, further more, a feedback scheduling system based on linear assignment algorithm was constructed, and in this system, processor utilization was assigned by feedback scheduler according to the control error, that is, real-time control tasks would get more processor utilization when its control error increases. Simulation results verified that the feedback scheduling system based on control error can improve the overall systems performance under limited computing resources
In this paper, the authors propose a scheme of multilevel parallel scheduling and traffic balancing, based on multiple constraints of packet switch network. By linear mapping tiny subsets of route space to network lin...
详细信息
In this paper, the authors propose a scheme of multilevel parallel scheduling and traffic balancing, based on multiple constraints of packet switch network. By linear mapping tiny subsets of route space to network links, we obtain the single solution and quantization formulation of split scheduling algorithm according to non-cooperative route scheduling strategy at single node. Furthermore, we also present the concept of link factor to descript the carrying capacity of links more accurately. The cyclic biding algorithm was designed to optimize the time-weighed property, which passes down the single solution to parallel controlling nodes and parallelizes the computation of IP header reorganization. It also was provided that the approximate method of computing the extended granularity of the controlling nodes
暂无评论